The Skuldelev Ships I


The Skuldelev Ships I

Topography, archaeology, history, conservation and display

Here follows the summary of: Crumlin-Pedersen, Ole & Olaf Olsen (eds) 2002: The Skuldelev Ships I. Topography, Archaeology, History, Conservation and Display. Ships and Boats of the North 4.1. Roskilde, 360 pages. ISBN 8785180467. Published by the Viking Ship Museum and the National Museum of Denmark.

This is the first volume of two which present the accumulated results of the work on the Skuldelev-find, a barrier from the 11th century in a channel in Roskilde Fjord, Denmark, containing five ships of varying type and provenance. The first volume presents the ships and the barrier in their historical and topographical context, as well as the conservation of the ships and their subsequent restoration and exhibition.

The first chapter covers the geological history of Roskilde Fjord (1.1-1.3), a knowledge of which is necessary for understanding the topographical situation at Skuldelev, where the ships were sunk in one of the channels in the fjord, Peberrenden. The ships had been scuttled at a place where a strong current passed over a natural barrier at the bottom of the channel, creating conditions for the deposition of a thick layer of oyster shells in the area. This barrier was part of the crest of a ridge formed during the melting phase of the last Ice Age. The ridge crosses the fjord in a north-south direction and it is still visible at Skuldelev Ås, on Kølholm and on Kalvøen at Frederikssund. The variations in the sea level here are discussed in Chapter 1.4 against the background of more recent finds and investigations and the chapter concludes that the sea level in the 11th century probably was the same, or at the most 0.5 m higher, than it is at present.

In the second chapter, the primary excavators of the find give a short account of the early history of the find (2.1), as well as the underwater investigations in 1957-59 and the excavation in 1962 (2.2-2.3). Traces have been found of several other barriers in the area (2.4) and samples taken from these suggest that they are contemporary with the Peberrenden barrier, but comprehensive industrial extraction of oyster shells for chicken feed have made it impossible to examine these finds more closely.

The third chapter relates the documentation methods employed in registering the ships' parts at the scale of 1:1 before they were conserved. The documentation made at this time subsequently formed the working-basis for the analyses of each individual ship's shape, size, original appearance, and phases of repair (3.1). In connection with this documentation a number of observations were made of traces left by tools and the cleaving of the timber; these have served as starting-points for the building of the Skuldelev full-scale reconstructions. Later, analyses were made of the caulking and other materials (3.2). The comprehensive dendrochronological analyses of timber from the ships are presented in Chapter 3.3, the results of which are described in Chapter 5, in relation to each individual vessel.

The fourth chapter gives a description of the difficult process of conserving the ships (4.1). The timbers have been conserved with polyethylene glycol (PEG 4000) in several different treatments: in hot baths, by freeze-drying from water or butanol, or in a few cases, by simply pouring the PEG over the wood. The next section (4.2) briefly presents the history behind the establishment of the Viking Ship Museum, followed by a description of the principles of restoration of the ships for exhibition in the museum. Supporting frames of L-shaped rails were built into the preserved parts of the ships. They support the extant hull parts and indicate the original lines of the ship in the areas where nothing is preserved. The problems involved with the care of the ships in the museum, and their long-term preservation are also discussed in relation to the experience gained this far.

The fifth chapter contains the detailed descriptions and analysis of the five ships. On the basis of the surviving wrecks and conclusions drawn about the age of the individual ships, their origin, function, and history are further contextualised in Chapter 6. The results of these individual analyses are briefly as follows:

Skuldelev 1 (5.1 and 6.3.2)
The ship is a sturdy cargo ship, originally 16.0 m long, built in western Norway about 1030. The dendroanalysis indicates that the ship's pine strakes derive from the Sognefjord area. On at least three occasions the ship was repaired with oak planks. The trees used for two of these repairs were evidently in both cases felled in southern Scandinavia (possibly the Oslofjord region and Skåne/Sjælland). The fore stem section of this ship has a specially full and rounded form, where the two topmost planks lean in over at the top ('tumble home'). This stem shape is also known from west Norwegian and Danish sketches and boat-models and it is linked with the ship-term knörr, knar, which is recorded from the Viking Age and the medieval period. The knar appears before 1000 as a warship but later as a cargo-ship on the North Atlantic and in the Baltic.

Skuldelev 1, with its cargo-carrying capacity of 20-25 tons, is smaller than other ship-finds of the same character (e.g. Hedeby 3), and is thus interpreted as an austrfararknörr, a knar for sailing to Denmark and into the Baltic. This is supported by the ship's find-place and by the repeated repairs carried out in eastern Norway and Denmark. Around 1030 there seems to have been a period of famine in Norway and there would have been a strong need for ships like this to pick up food supplies in Denmark.

Skuldelev 2 (5.2 and 6.4)
The ship is an approx. 30 m-long longship, built in the region of Dublin in Ireland but within the Nordic shipbuilding tradition. The oak trunk from which the keelson has been shaped was felled in May-June 1042, and the ship was therefore probably launched later that year. In the 1060s the ship was repaired with planks from an oak that was also felled in the British Isles. Only about 20% of the ship is preserved, but the parts that are extant are from various places in the hull, so it has been possible to calculate the size of the ship within narrow limits. It was a warship for about 60 rowers/warriors as well as several others, such as a lookout, a steersman, etc., so that a complete crew may have totalled up to 80 men. This ship, like the others of the find, would have carried mast and sail, and the shape of the hull is well suited to navigation on the open sea. The correct, contemporary term for this type of ship was likely to be skeið.

Skuldelev 2 probably sailed its first ca 25 years around the Irish Sea, taking part in the mid-eleventh-century military operations that Viking towns in Ireland, in particular Dublin, were involved in. The reason for this ship's journey to Roskilde in Denmark may be related to the events following William the Conqueror's invasion of England in 1066. The King of England, Harold Godwinson, was killed and his sons and daughter had to take flight. His sons came to Dublin from where they attacked the south-western part of England in 1068. According to the Danish chronicler Saxo, two of Harold's sons and a daughter came to Denmark to their uncle, King Sven Estridsson. In 1069 new attacks were made on England from Dublin, and the same year King Sven of Denmark sent a large fleet. As Skuldelev 2 was built in 1042 and repaired after the middle of the 1060s in the Irish Sea region, and sank in Roskilde Fjord in the 1070s, the possibility exists that this longship was the one in which Harold's sons and daughter travelled from Dublin to Roskilde on a diplomatic mission in 1068-69, with a view to restoring Danish or Anglo-Saxon rule in England by a co-ordinated military operation.

Skuldelev 3 (5.3 and 6.2.1)
The ship is a small transport and cargo ship, originally 14.0 m long with slender lines that display consistently good workmanship. It was primarily a sailing-ship but it could be rowed at positions before the mast. The ship was built about 1040 of oak from Denmark. Analyses of the shape of the fore-stem have indicated some of the principles that were employed when building without the aid of plans the complicated and well-proportioned ships and boats of that period. The details show that the ship-type is based on earlier traditions, utilised to build ships of this size with seven strakes on each side, while for this ship it was necessary to employ an eighth strake to achieve the same sheer-height. Consequently, it must have been difficult to find sufficiently broad planks to keep to the same, earlier standard.

Skuldelev 3 was probably of the type referred to as a byrding or skude. The vessel had a cargo-carrying capacity of 4-5 tons and is to be considered a small cargo ship, primarily for the transport of the owner's personal produce, or a transport ship for a large estate or the like. It may also occasionally have functioned as a vessel for a travelling merchant acting on behalf of a local chieftain.

Skuldelev 4 (5.4) is not an independent ship-find but part of the starboard side aft of Skuldelev 2 and it is therefore discussed in the analysis of that ship.

Skuldelev 5 (5.5 and 6.2.2)
The ship is a smallish warship, originally 17.3 m long and adapted for rowing by 26 men. The ship was built in Denmark around 1030 and its bottom was constructed of new oak timbers, but planks of pine and ash from two different ships were re-used in the upper part of the sides of the ship. In the uppermost strake, not all the oarports from the original ship could be re-used for rowing Skuldelev 5, since the distance between the oars was different. Many of the holes had therefore to be covered over and new holes cut beside these. The ship was obviously built by an experienced boatbuilder. It is, however, the extensive use of second-rate materials and a relaxed attitude to the aesthetic qualities that characterise the construction of Skuldelev 5, in contrast to other ships of this period. As a result of to its long working life, Skuldelev 5 is heavily worn both internally and externally. There are many repairs, the last of which were done shortly before the sinking of the ship at a time when the wood in the keel was mouldering.

With its 13 pairs of oars, Skuldelev 5 represents the smallest class of longships and it was probably referred to as a snekke. This word forms part of about 100 place-names around the coast of Denmark, often assigned to places where longships could be laid up out of sight of passing ships. The character of the ship is hardly indicative of a prestige vessel for the personal use of kings or chieftains. Such vessels, as for example, Hedeby 1 and Roskilde 6, were normally built to a considerably higher standard. The existence of Skuldelev 5 may, on the other hand, be understood by reference to the regulations in the Gulating Law of western Norway, containing regulations concerning the building and use of the local leiðang-ships. The local farmers had, for example, the duty of providing materials for the building of the ship and it would seem that they tried to cut as many corners as possible when doing this. Against this background Skuldelev 5 provides circumstantial evidence of a coastal defence in Denmark, already around 1030, based on ships built and manned according to the same principles as in the leiðang-arrangement which is known a century later from western Norway.

Skuldelev 6 (5.6 and 6.3.3)
The ship is the smallest ship of the find, originally about 11.2 m long, built in the same region in western Norway and at the same time, ca 1030, as Skuldelev 1. The vessel was originally built in six strakes of pine planks, supported in the sides by sturdy beams (bitis) and knees. Later, while the ship was still in a region where there was access to pine forests, it was converted by the addition of a broad seventh strake. Skuldelev 6 has a modest keelson with mast-step but in its original form it was probably equipped with rowlocks or tholepins for six oars on each side. As in the case of Skuldelev 1, the repairs to Skuldelev 6 were carried out with oak planks but the provenance of these planks has not been identified.

It is probable that Skuldelev 6, in its first form, was built for use in western Norway for fishing and hunting seals and small whales in the fjords and along the coast. Its full lines, strong frames, and ability to carry several men would have been important elements in the hunting and landing of these animals. After its conversion, the vessel changed character. It would have been able to have, at the most, a few oars fore as in Skuldelev 3. The cargo-carrying capacity of Skuldelev 6 increased, but at the same time the vessel would no longer have been so versatile. Its function would have been restricted to that of a transport ship either locally or further afield, as shown by its finding-place. It is likely that Skuldelev 6 had the same business in Denmark as Skuldelev 1, and perhaps they even came to the region at the same time. Both ships could also have sailed regularly on the route to Sjælland, however.

The final section (6.5) presents the combined evidence from all the analyses in order to date the phases of the barrier and to relate these to events known from contemporary written sources. It is clear that the ships were sunk on two different occasions, first Skuldelev 1, 3, and 5, and then after an interval of a few years, Skuldelev 2, '4', and 6. Subsequently the barrier was extended with the erection of posts and fascines but without the use of more ships. No traces have been found to suggest that the barrier in one or more of its phases was constructed as an enemy action against Roskilde. On the contrary, the barriers in Peberrenden and the other channels seemed to have formed a major complex, erected and expanded on several occasions with a view to warding off enemy attacks. On the basis of the dates of construction of the ships and an estimate of their life-span in active service, it can be assumed that the first phases of the Peberrenden barrier was built in the 1060s, and a second phase in the 1070s, and the third phase with several poles post-dates this.

In analysing the few textual sources about Danish history in the 11th century, it is tempting to point to King Harald Hardrada’s ravaging of the Danish coast until 1064 as a possible reason for the building of the first stage, while the second stage may reflect the struggles for the throne after the death of Sven Estridsson in 1074 or 1076. The barrier, after these two stages of construction, evidently prevented King Niels' fleet from sailing all the way to Roskilde during the conflict with Erik Emune before the Battle at Værebro River in 1133. On the basis of a single 14C date, it is assumed that at least some of the posts and fascines of the third phase are indicative of events in 1288, following the murder of King Erik Klipping, which included sea-borne raids from Norway into the Isefjord complex, of which Roskilde Fjord is a part.

The forthcoming Volume II of the present monograph will analyse the ships functionally. Additionally, details will be presented of the experience and knowledge that have been gained as a result of the experimental analyses of the ships in relation to the building and sailing-trials of full-scale reconstructions of the ships.

Share or add bookmark

Print Tip en ven

Read more

» By "The Skuldelev Ships I" in the museum webshop

» Publications by the Viking Ship Museum 

The Skuldelev Ships I. Topography, Archaeology, History, Conservation and Display. Ships and Boats of the North 4.1. Edited by Ole Crumlin-Pedersen & Olaf Olsen. Roskilde 2002. Published by the Viking Ship Museum and the National Museum of Denmark. Photo Werner Karrasch

Vikingeskibsmuseet: Vindeboder 12 . DK-4000 Roskilde | Tlf.: +45 46 300 200 | museum(at) |


logo uses a number of cookies to improve your user experience. Cookies are used to generate statistics concerning the use of and its self-service solutions, so that we can continue to make improvements to them. The cookies from that are stored on your computer can be deleted at any time.

» Read more about cookies on the


Accept cookies